CELLULAR ELECTROPHYSIOLOGY OF THE HEART 



263 



equimolar quantities of choline+. If NaCl was re- 

 placed by sucrose, the effects — although in the same 

 direction — were much smaller (4). The difTerences are 

 probably due both to an appreciable contribution of 

 Cl^ to repolarization and to an increase in §,< re- 

 sulting from an acetylcholine-like action of the choline 

 on the ventricle. These results further indicate that 

 Na+ current maintains the plateau. With substitu- 

 tion of choline, Iap is directly proportional to [Na+]o, 

 projecting to zero at an [Na+]„ of about 15 per cent 

 of normal (4). 



An increase in [K+]o shortens tAp because over- 

 shoot is reduced and, in frog ventricle, because the 

 rate of third phase repolarization increases (3, cf. 

 fig. 18). The latter change is not found in all tissues 

 [fig. 19; (130)]. It would be expected that, aside 

 from the effects of the reduced Sr, an increase in 

 [K+]o would have little effect on early repolarization, 

 since the repolarizing current flows outwardly and 

 should be little afTected by [K+]o if the independence 

 principle holds. However, the shortening of the 

 action potential is clearly not due entirely to a re- 

 duction in overshoot. Weidmann (130, 131) found 

 that a slug of K+ injected into the coronary circula- 

 tion of a turtle ventricle during the plateau phase 

 produced early repolarization (fig. 19), indicating 

 a direct effect of K+ on membrane properties. He 

 suggested three possible mechanisms of this action : 

 /) The high [K+]o increases the rate of Na"*" pump- 

 ing; this explanation seems unlikely, since the pump 

 is probably neutral. 2) The gK depends directly on 

 [K+]o; this possibility is supported by the increase 

 in repolarization rate with [K+Jo in frog ventricle 

 (fig. 18) and the radioactive potassium studies of 

 Carmeliet (9), but is contradicted by the slower 

 repolarization rate of the action potential shortened 

 by a slug of K+ (fig. 19). 5) An increase in [K+]o 

 decreases g^,,; a decreased g^;, would explain the 

 records in figure 19 but not those in figure 18. An- 

 other possibility is that gci depends on [K+]o but 

 recent experiments (8, 73) indicate that C\ carries 

 about 20 per cent of the repolarizing current. Thus 

 no definite explanation of the effects of [K+]o on 

 the repolarization can be selected; and it is quite 

 possible that all cardiac tissues do not react to K+ in the 

 same manner. 



The large effects of the interval between two suc- 

 cessive stimuli on the duration and the near lack of 

 effects on the shape of the action potential are shown 

 in figure 20. Action potentials evoked successively 

 later, during and after repolarization, have pro- 

 gressivelv greater amplitudes and durations, 



e-Er 



(mV) 

 100 r 



50 - 



C- 



12 3 4 

 TIME (Sec) 

 FIG. 19. Shortening of the cardiac action potential brought 

 about by suddenly raising [K+Jo during the plateau. Prepa- 

 ration was tortoise ventricle perfused via the coronaries. The 

 longer action potential is a control, evoked by a stimulus 

 given at zero time. The shorter action potential, also evoked, 

 was recorded when a "slug" of high [K+] solution was added 

 to the perfusion fluid at the time indicated by the pulse in the 

 lower record. Note that increased [K^'^lo has a repolarizing 

 action during the plateau and a depolarizing action at rest. 

 Ordinate: change in transmembrane potential (£ — Sr) in 

 millivolts; abscissa: time after stimulus in seconds [After 

 Weidmann (131).] 



but the duration is still shorter than normal long 

 after the threshold returns to normal. Carmeliet & 

 Lacquet (10) have shown that, when stimuli are 

 applied rhythmically to frog ventricle, the relationship 

 between tAp and the stimulus interval (tj) is accurately 

 exponential, the time constant of recovery of dura- 

 tion being just over 1 sec. This behavior has sur- 

 vival value since it provides that the diastolic and 

 systolic periods be shortened or lengthened together, 

 thus insviring appropriate emptying and filling 

 times. This shortening is a characteristic property of 

 cardiac muscle and explanation of it must be in- 

 cluded in any hypothesis of repolarization. The tAp,ts 

 relationship arises from time dependent processes. 

 In fact its behavior is quite similar to the activation- 

 inactivation process for g^;, on a slow time scale. 



SUPERIMPOS.'>iBILlTY OF ACTION POTENTI.'^LS. Bradv & 



Woodbury (4) have pointed out an interesting char- 

 acteristic of the action potentials evoked at different 

 stimulus rates: their third phases can be super- 

 imposed by shifting them in lime. This phenomenon 

 is illustrated in figure 20B, where the action po- 

 tentials ( I to 5) of figure 20.-1 have been shifted in 

 time so that their third phases coincide. The third 

 phases superimpose quite well except that AP i is 

 steeper than the others just before repolarization is 

 completed. The variation between the second phases 

 of the shortened action potentials is scarcely more 

 than the variations of the control action potentials 

 in figure 20.-^. 



